Article pubs.acs.org/jmc
New Serotonin 5‑HT1A Receptor Agonists Endowed with Antinociceptive Activity in Vivo Margarita Valhondo,†,∥ Isabel Marco,†,⊥ Mar Martín-Fontecha,† Henar Vázquez-Villa,† José A. Ramos,‡ Reinhard Berkels,§ Thomas Lauterbach,§ Bellinda Benhamú,*,† and María L. López-Rodríguez*,† †
Departamento de Química Orgánica I, Facultad de Ciencias Químicas, Universidad Complutense de Madrid, E-28040 Madrid, Spain Departamento de Bioquímica y Biología Molecular, Facultad de Medicina, Universidad Complutense de Madrid, E-28040 Madrid, Spain § UCB Pharma GmbH, Alfred-Nobel-Strasse 10, 40789 Monheim, Germany ‡
S Supporting Information *
ABSTRACT: We report the synthesis of new compounds 4−35 based on two different openings (A and B) of the chromane ring present in the previously identified 5-HT1A receptor (5-HT1AR) ligand 3. The synthesized compounds were assessed for binding affinity, selectivity, and functional activity at the 5-HT1AR. Selected candidates resulting from B opening were also evaluated for their potential antinociceptive effect in vivo and pharmacokinetic properties in vitro. Analogue 19 [2-(4-{[2-(2-ethoxyphenoxy)ethyl]amino}butyl)tetrahydro-1H-pyrrolo[1,2-c]imidazole-1,3(2H)-dione] has been characterized as a high-affinity and potent 5-HT1AR agonist (Ki = 2.3 nM; EC50 = 19 nM). Pharmacokinetic studies indicated that compound 19 displays a good metabolic stability in human liver microsomes (t1/2 ∼ 3 h and CLint = 3.5 mL/min/kg, at 5 μM), and a low level of protein binding (25%, at 5 μM). Interestingly, 19 (3 mg/kg, ip, and 30 mg/kg, po) caused significant attenuation of formalin-induced behavior in early and late phases of the mouse intradermal formalin test of pain, and this in vivo effect was reversed by the selective 5-HT1AR antagonist WAY-100635. Thus, the new 5-HT1AR agonist identified in this work, 19, exhibits oral analgesic activity, and the results herein represent a step toward identifying new therapeutics for the control of pain.
■
produces analgesia as a “first-order” effect and hyperalgesia as a “second-order” effect. The chronic treatment with opiods attenuates the first-order analgesia and strengthens the secondorder hyperalgesia. According to this concept, it has been proposed that 5-HT1AR activation has two effects that are opposite from those produced by opioids: pain as a first-order effect but also hypoalgesia as a second-order effect; with chronicity, this second-order hypoalgesia counteracts the firstorder pain and remarkably develops into increasingly powerful analgesia.26 This suggests that whereas repeated morphine administration leads to a loss of analgesic activity (tolerance) and even to hyperalgesia, repeated administration of a 5-HT1AR agonist could produce the opposite effect, i.e., a progressive loss of acute hyperalgesia and an increase in analgesia.27 These considerations stimulated pharmaceutical drug discovery that aimed to identify 5-HT1AR agonists with high selectivity. In particular, the potent and selective 5-HT1AR agonist F-13640 [befiradol, 1 (Chart 1)] demonstrated analgesic activity in animal models of acute and chronic pain comparable to those of large doses of opioid painkillers, but with fewer and less prominent side effects, as well as little or no development of tolerance with repeated use.27,28 These studies represented a proof of concept that the 5-HT1AR is involved in the regulation of nociceptive
INTRODUCTION Serotonin (5-hydroxytryptamine, 5-HT) mediates a plethora of physiological effects through at least 14 receptor subtypes, all but one belonging to the G-protein-coupled receptor (GPCR) or seven-transmembrane-spanning (7TM) receptor family. Defined on the basis of molecular, pharmacological, and functional criteria, 5-HT receptors have been classified into seven discrete subfamilies (5-HT1 to 5-HT7),1 and their study for almost three decades has provided researchers with many targets for drug action. The 5-HT1A receptor (5-HT1AR) was the first 5-HT subtype to be fully sequenced,2 and it has also been the most extensively studied among serotonin receptors. Indeed, a large number of selective or multitarget 5-HT1AR ligands have been developed so far with the aim of identifying new drugs, and this receptor still represents an attractive target for drug discovery.3−7 In particular, 5-HT1AR agonists and partial agonists have long shown to be clinically effective in the treatment of anxiety, depression, and psychosis.8−15 More recently, the neuroprotective properties observed for some 5-HT1AR agonists have suggested their utility for the treatment of ischemic stroke,5,16−22 and these agents have also been proposed as levodopa adjuvants in the alleviation of dyskinesia in Parkinson’s disease.23,24 Moreover, one of the most attractive therapeutic potentials that has recently emerged for 5-HT1AR agonists is that they may rival the opioids in pain relief therapy.25 It is known that in nociceptive systems any input causes two effects that are opposite in sign. Activation of opioid receptors © XXXX American Chemical Society
Received: May 23, 2013
A
dx.doi.org/10.1021/jm400766k | J. Med. Chem. XXXX, XXX, XXX−XXX
Journal of Medicinal Chemistry
Article
pharmacokinetic properties in vitro. In particular, compound 19 was characterized as a high-affinity and potent 5-HT1AR agonist that exhibited oral analgesic activity in the mouse intradermal formalin test and a good pharmacokinetic profile.39
Chart 1. Structures of 5-HT1AR Agonists Endowed with Antinociceptive Activity
■
RESULTS AND DISCUSSION Synthesis. Target compounds 4−7 based on A opening of the chromane ring in ligand 3 (Chart 2) were obtained by nucleophilic substitution of 2-(4-bromobutyl)tetrahydro-1Hpyrrolo[1,2-c]imidazole-1,3(2H)-dione (36) with the appropriate commercial ω-(o-methoxyphenyl)alkylamines or amine 37 in acetonitrile, which was prepared starting from 3-(2methoxyphenyl)propanoic acid (Scheme 1). Final compounds 8−35 resulting from B opening of the chromane ring in ligand 3 (Chart 2) were synthesized starting from halogenated intermediate 36, by reaction with appropriate 2-aryloxyethylamines 38, or by alkylation of 2-(4-aminobutyl)tetrahydro-1Hpyrrolo[1,2-c]imidazole-1,3(2H)-dione (39) with corresponding 2-aryloxyethyl bromides 40, as described in Scheme 2. Starting compound 39 was obtained by N-alkylation of hydantoin 41 with 4-chlorobutanonitrile in the presence of sodium hydride, followed by catalytic hydrogenation of intermediate 42. Williamson reaction of appropriate phenols with 1,2-dibromoethane or 2-chloroacetamide in the presence of potassium carbonate and catalytic potassium iodide in refluxing 2-butanone afforded 2-aryloxyethyl bromides 40a−t or intermediate 2-aryloxyacetamides 43a−g. Subsequent reduction of 43a−g with the borane dimethyl sulfide complex in diglyme yielded 2-aryloxyethylamines 38a−g. Intermediate amine 38h was prepared by Mitsunobu reaction starting from 8-hydroxyquinoline and N-(2-hydroxyethyl)phthalimide, followed by phthalimide cleavage. Binding Affinities. New synthesized compounds 4−35 were converted to hydrochloride salts, and their affinities were evaluated at the 5-HT1AR by radioligand binding assays (see the Supporting Information for details). High-affinity 5-HT1AR ligands (Ki < 50 nM) were also assayed for selectivity over serotonin 5-HT2A, 5-HT3, 5-HT4, and 5-HT7 receptors. Tables 1 and 2 show the calculated inhibition constants (Ki)40 for target compounds 4−35 obtained from A and B openings of the chromane ring in ligand 3 (Chart 2). Clearly, A opening of chromane ring leads to a marked decrease or loss of affinity for the 5-HT1AR, affording inactive or poorly active derivatives [Ki(4−7) ≥ 74 nM vs Ki(3) = 1.23 nM] (Table 1). On the other hand, compound 9 that results from B opening of chromane ring in compound 3 maintained a high affinity for the receptor [Ki(9) = 16.8 nM] (Table 1). Therefore, we decided to study the influence of several substituents at different positions of the aromatic ring in compounds 8 and 10−35. From data in Table 1, in general ortho and meta positions are more favorable for 5-HT1AR affinity than the para position, regardless of the nature of the substituent. Thus, all para-substituted derivatives are inactive, including those with substituents that provide high 5-HT1AR affinity [Ki(9) = 16.8 nM and Ki(10) = 30 nM vs Ki(11) = 742 nM; Ki(16) = 4.8 nM and Ki(17) = 34 nM vs Ki(18) > 1000 nM]. Also, the presence of a substituent in the ortho position is in general more favorable than that in the meta position (see, for example, compounds 9 vs 10, 13 vs 14, and 16 vs 17). Electron-withdrawing substituents seem to be detrimental for affinity even in the preferred ortho position [Ki(24) > 1000 nM and Ki(27) = 510 nM]. Accordingly, compounds bearing ortho-alkoxy groups displayed the highest 5-HT1AR affinities [Ki(16) = 4.8 nM, and Ki(19) = 2.3 nM]. With regard to selectivity, in general, the new
transmission. Therefore, 5-HT1AR activation was proposed as a new molecular approach to the treatment of acute and chronic, nociceptive and neuropathic pain states. Indeed, the concept has been extended to other 5-HT1AR agonists such as Lu AA21004 [vortioxetine, 2 (Chart 1)] that has shown analgesic activity in the formalin test.29,30 Given the complexity of the mechanisms involved in pain modulation, it is not surprising that the application of single analgesic agents is not always effective in diverse painful conditions such as chronic pain syndromes. Thus, it is becoming clear that the combination of different mechanisms, which improves efficacy with reduced toxicity, is necessary for the reliable pharmacotherapy of pain. At present, there is evidence that the 5-HT1AR may be used as a target in the search for new pharmacological approaches in the augmentation of analgesia. The ongoing clinical trials with compound 131 have certainly reinforced the interest of 5-HT1AR activation for the treatment of pain conditions. In this context, our goal is the development of a 5-HT1AR agonist as an antinociceptive agent. Over the past several years, our efforts in the search for new and selective 5-HT1AR ligands have led to the discovery of several arylpiperazine-based 5-HT1AR agonists endowed with anxiolytic properties.3,32−37 More recently, we have developed new nonpiperazine derivatives as second-generation 5-HT1AR agonists exhibiting a neuroprotective effect against ischemic cell damage.22,38 In these series, the modification of the different structural moieties of the molecule led to highest-affinity ligand 3 (Chart 2), which was characterized as a potent 5-HT1AR agonist Chart 2. Designed Compounds Based on A (4−7) or B (8−35) Opening of the Chromane Ring Present in the Previously Identified High-Affinity 5-HT1AR Ligand 3
(Ki = 1.23 nM; EC50 = 16.3 nM). In this work, we aimed to generate structural variants of ligand 3, focusing our attention on mimicking the chroman-2-ylmethyl group. In the new compounds, two different openings (A and B) of the chromane ring were considered (Chart 2). The synthesized compounds have been assessed for binding affinity, selectivity, and functional activity at the 5-HT1AR. Selected candidates have also been evaluated for their potential antinociceptive effect in vivo and B
dx.doi.org/10.1021/jm400766k | J. Med. Chem. XXXX, XXX, XXX−XXX
Journal of Medicinal Chemistry
Article
Scheme 1. Synthesis of Target Compounds 4−7a
a
Reagents and conditions: (a) LiAlH4, Et2O, rt, 20 h, 48%; (b) SOCl2, toluene, pyridine, rt, 24 h, 77%; (c) KCN, KI, DMSO, rt, 3 h, 70%; (d) LiAlH4, THF, rt, 3.5 h, 56%; (e) CH3CN, 60 °C, 15 h, 30−52%.
Scheme 2. Synthesis of Target Compounds 8−35a
a Reagents and conditions: (a) CH3CN, 60 °C, 15 h, 21−69%; (b) NaH, DMF, 110 °C, 4 h, 85%; (c) H2, PtO2, EtOH, rt, 15 h, 35%; (d) Br(CH2)2Br, K2CO3, KI, 2-butanone, Δ, 48 h, 30−67%; (e) ClCH2CONH2, K2CO3, KI, 2-butanone, Δ, 48 h, 53−98%; (f) BH3·S(CH3)2, diglyme, 90 °C, 16 h, 61−88%; (g) N-(2-hydroxyethyl)phthalimide, Ph3P, DEAD, THF, Δ, 15 h, 50%; (h) N2H4·H2O, CH3COOH, EtOH, Δ, 3 h, 33%.
equilibrated at 37 °C with NADPH and MgCl2 at concentrations of 1 and 5 μM; RLM and HLM preparations were added to initiate the reaction, and aliquots were withdrawn and quenched at different times (see the Experimental Section for details). The percent of parent compound remaining in the incubation mixture was determined by high-performance liquid chromatography and mass spectrometry (HPLC−MS) and plotted as a function of time. The half-life time (t1/2) was determined from the slopes of the peak areas over time, which was then used to calculate the intrinsic clearance (CLint).41 From the results listed in Table 4, microsomal metabolism was found to be more favorable for HLMs than for RLMs. The data indicated that all assayed compounds exhibited high rates of metabolic degradation in rats, with t1/2 values of 8−13 min and CLint values of >100 mL/min/kg (Table 4). However, good rates were observed in human microsomes with t1/2 values of 2−4 h and CLint values of
